Culture 3 min read

Researchers Develop a New AI System That Offers Fashion Advice

Prostock-studio / Shutterstock.com

Prostock-studio / Shutterstock.com

What if you could turn to your smartphone for fashion advice? Imagine if an AI system could use your device’s camera to scan your outfit and tell you how to make it look better.

Sounds a bit too futuristic, right? Well, not as much as you may think.

A team of computer scientists from the University of Texas teamed up with researchers from Cornell tech, Georgia tech, and Facebook AI research to develop a system that gives style advice. The artificial intelligence system can look at the photo of an outfit and suggest ways to help you keep up with the current popular style.

For example, it could suggest you pick a sleeveless top or a longer jacket to make your outfit look more fashionable. The tool could also tell you how to match a colored tie to your suits and shirts.

Speaking about the project, a professor of computer science and visual recognition in AI expert, Kristen Grauman said:

“We thought of it like a friend giving you feedback. It’s also motivated by a practical idea: that we can work with a given outfit to make small changes. So it’s just a bit better.”

The researchers are calling the artificial intelligence system Fashion++.

The tool uses visual recognition system to analyze elements of an outfit, such as the pattern, color, texture as well as its shape. It considers where edits will have the most impact, and then offers other possible choices of outfits to users.

Creating an Artificial Intelligence System to Give Fashion Advice

The researchers collected over 10,000 images of publicly shared outfits on online fashion sites to train the AI system.

According to graduate student Kimberly Hsiao, finding images of fashionable outfits was relatively easy. The challenge was finding the unfashionable ones.

So, she got creative and mixed images of fashionable outfit to create a less fashionable outcome. Then, the team trained the system on what not to wear.

Hsiao noted:

“As fashion styles evolve, the AI can continue to learn by giving it new images, which are abundant on the internet.”

As great as it is for AI to give fashion advice, it’s not free from bias datasets.

The researchers noted that Fashion++ struggled to recognize vintage looks as stylish because the training images came from the internet, which gained extensive use in the 1990s. Also, since most of the training images were from North America, styles from other parts of the world rarely show up on the system.

Finally, many images of fashionable clothing in the tool appear on models with similar shapes. Meanwhile, bodies come in various shapes and sizes, which in turn influences the ideal choice of outfit.

“We are examining the interaction between how a person’s body is shaped and how the clothing would suit them,” Grauman said. “We’re excited to broaden the applicability to people of all body sizes and shapes by doing this research.” 

Read More: AI Assessment System Helps Employers With Hiring Process

First AI Web Content Optimization Platform Just for Writers

Found this article interesting?

Let Sumbo Bello know how much you appreciate this article by clicking the heart icon and by sharing this article on social media.


Profile Image

Sumbo Bello

Sumbo Bello is a creative writer who enjoys creating data-driven content for news sites. In his spare time, he plays basketball and listens to Coldplay.

Comments (0)
Most Recent most recent
You
share Scroll to top

Link Copied Successfully

Sign in

Sign in to access your personalized homepage, follow authors and topics you love, and clap for stories that matter to you.

Sign in with Google Sign in with Facebook

By using our site you agree to our privacy policy.